The present invention relates to magnetic resonance imaging, and more particularly to a method and apparatus for performing functional magneti resonance imaging (fMRI) in conscious animals.
Human studies utilizing fMRI have advanced our understanding of the regional and functional interplay between populations of neurons serving sensory, integrative and motor functions. Changes in neuronal activity are accompanied by specific changes in hemodynamics such as cerebral blood flow, cerebral blood volume, and blood oxygenation. Functional MRI has been used to detect these changes in response to visual stimulation, somatosensory activation, motor tasks, and emotional and cognitive activity. When the brain is activated by any of these conditions, the blood flow and delivery of oxygen to the active regions the tissue oxygen uptake resulting in an increase in blood oxy-hemoglobin (Hb02) content. The susceptibility difference between diamagnetic oxy-hemoglobin and paramagnetic deoxy-hemoglobin (Hb) creates local magnetic field distortions that affect the processional frequency of the water protons. The consequential change in magnetic resonance (MR) signal intensity which is proportional to the ratio of Hb02 to Hb. These signal-intensity alterations related to blood oxygenation are termed the BOLD (blood oxygenation-level-dependent) effect. The voxels in paramagnetic Hb content is decreased are illuminated in the image.
While most work on fMRI has been done in humans, it has been difficult to use this technology in conscious animals because of motion artifact. As a result, most studies to date have been limited to animals which are typically anesthetized in order to minimize this problem of motion artifacts. The low level of arousal during anesthesia either partially or completely suppresses the fMRI response and has impeded fMRI application to the more physiologically relevant functions that have been noted in humans.
Since image resolution is a salient feature of fMRI, precautions to ensure improved image quality with minimized head movements are essential. In addition to head movement, it has been observed that any motion outside the field of view can obscure or mimic changes in signal.
Another, equally significant component for achieving high temporal and spatial image resolution is the generation of radio frequency (RF) magnetic fields. The RF field pulses are transmitted to flip protons into the transverse plane of the main direct current (DC) magnetic field. As these protons precess and relax back into the longitudinal plane of the main magnetic field they emit RF magnetic field signals. The electrical assemblies capable of sending and receiving RF signals are called RF probes, coils, or resonators. Ideally, a RF coil used for magnetic field transmission creates a large homogenous area of proton activation at a very narrow bandwidth center around the proton resonance frequency with minimal power requirements. An RF coil used for receiving covers the largest region of interest within the sample at the highest signal-to-noise ratio (SNR). RF coils are either volume coils or surface coils. A volume coil has the advantage of both sending and receiving RF signals from large areas of the sample. However, signal-to-noise ratio is compromised because a large spatial domain contributes to the RF signal, resulting in additional noise and thereby obscuring the RF signal from the region of interest. A surface coil has the advantage of improved SNR due to its close proximity to the sample. Unfortunately, a surface coil is ill suited for RF energy transmission owing to the fact that only a small proton area can be activated. Two criteria are sought in the design of superior coil performance for high field animal studies. First, the coils must be as efficient as possible. Transmission efficiency is increased by reducing the resistive coil losses through appropriate arrangement of conductors, the use of a shield, and the employment of low loss dialectric materials. By using a separate surface coil in proximity over the desired field of view (FOV) or region of interest, the reception efficiency of the acquired NMR signal is further increased. In imaging, spatial and temporal resolutions are proportional to SNR.
The second criterion to be met for a volume coil, is the uniformity or homogeneity over a desired FOV in the animal sample. To achieve both homogeneity and efficiency for volume coils of Larmor wavelength dimensions, further improvements are required. Conventional state-of-the-art birdcage coil designs will not resonate at these dimensions.
So-called transversal electromagnetic (TEM) resonator designs have shown promise for high-frequency, large volume coil applications for humans. However, these TEM designs must be improved upon for the highest frequency and animal applications allowed by present and future magnets, e.g; for the 9.4T, and the 11.74T, magnets presently being built for laboratory animal studies.
Increased SNR is sought by making NMR measurements at higher magnetic, or Bo, fields. Main magnetic field strength is, however, only one of several parameters affecting the MR sensitivity. RF coil and tissue losses can significantly limit the potential SNR gains realized at high fields. SNR (and reciprocal transmission efficiency) will suffer when the coil's ohmic resistance, radiation resistance, coupled tissue losses, RF magnetic field and angular frequency are not optimized.
Tissue losses increasingly impact SNR at higher frequencies. These conductive and dielectic losses represented are limited in practice by using local surface coils, or volume coils efficiently coupled to a region of interest. In addition to tissue loading, RF losses in the coils themselves become significant at higher frequencies. The RF coil loss increases with frequency as do the resistive losses in the coil RC, which increases with the square root of the angular frequency, and the losses from radiation resistance, which increases as at the fourth power of the angular frequency. The radiation losses also increase as the coil size increases as S2, where S is the area bounded by the coil.
From the above, it is apparent that radiative losses to the sample and environment, as well as conductive losses to the load of a coil become severe to the point of limiting and eventually degrading the SNR gains otherwise expected at higher magnetic field strengths. Physically, as a coil is increased in dimension and/or frequency, its electrical circuit length increases, the coil ceases to behave like a “coil”(RF field storage circuit) and begins to behave more like an “antenna” (RF field energy radiator).